fcs2InteractiveLikelihood: Interactive FCS2 model likelihood
In aquaMetrics/fcs2: Fisheries Classification Scheme 2 For SNIFFER

Description Usage Arguments Warning See Also Examples

View source: R/fcs2InteractiveLikelihood.R

Creates a plot of the FCS2 model likelihood with interactive controls that allow the model parameters to be adjusted. The likelihood represents the probability of obtaining a single fish count and the controls allow you to see how these probabilities change with the parameters. The function can visualise either the zero-inflated negative binomial (ZINB) distribution (see zinbinom) of the single catch C in the original EA FCS2 model or the ZINB distribution of the total catch T over all passes from the multiple-pass FCS2 model.

fcs2InteractiveLikelihood(
  multipass = FALSE,
  eqr = FALSE,
  boundaries = NULL,
  r.limit = c(0.1, 10),
  mu.limit = c(0, 10),
  rho.limit = c(0, 1),
  a.limit = c(0, 10),
  d.limit = c(1, 7),
  q.limit = c(0, 1),
  c.limit = c(0, 20)
)

`multipass`	If `FALSE` (the default), the ZINB likelihood of the single catch C from the original EA FCS2 model is given. If `TRUE`, the ZINB distribution of the total catch T over all passes from the multiple-pass FCS2 model is given. The true likelihood for this model is the multivariate ZINM distribution (see `fcs2:::zinmultinom`).
`eqr`	whether to display the EQR value.
`boundaries`	a vector of length 4 giving the EQR boundaries separating the classes Bad, Poor, Good, Moderate and High. These are used only to colour the plot with Bad red and High blue. If `NULL` (default), the probability that defines the single EQR is coloured blue.
`r.limit, mu.limit, rho.limit, a.limit`	vectors of giving the lower and upper limits for the interactive controls for the shape parameter r, the abundance μ, the prevalence rho and the survey area a respectively.
`d.limit, q.limit`	vectors of giving the lower and upper limits for the interactive controls for the number of runs d and the catch probability q respectively. These are used only if `multipass = TRUE`.
`c.limit`	vector of giving the lower and upper limit for the interactive control for the single catch. This is used only if `eqr = TRUE`.

This function requires the additional package rpanel for producing interactive controls.

fcs2InteractivePrediction which plot predictions from a fitted FCS2 model with interactive controls to show how the predictions change when covariates are varied.

## Interactive likelihood of total catch with EQR given against class boundaries

# only run example if R is running interactively
if (interactive()) {
    # define WFD class boundaries
    boundaries <- c(0.001, 0.01, 0.25, 0.5)

    # create interactive plot
    fcs2InteractiveLikelihood(multipass=TRUE, eqr=TRUE, boundaries=boundaries)
}